Machine learning kinetics from molecular dynamics data
Abstract
Most molecular transitions occur on time scales far beyond direct molecular dynamics simulations. The committor, the probability that a configuration reaches a product state before a reactant state, is a central kinetic statistic, providing a mechanism-independent reaction coordinate and a foundation for transition path theory and the calculation of rates. This review surveys modern approaches for estimating the committor and related kinetic statistics from molecular simulations, with an emphasis on self-supervised methods that learn solutions of their defining dynamical equations rather than relying on labeled shooting data. We develop a common operator viewpoint connecting generator-based partial differential equations, variational principles, Markov state models, dynamical Galerkin approximation, and neural networks. Empirical and theoretical evidence points to the efficiency of these methods. We provide theoretical and practical guidance for realizing their full potential in applications, including strategies for treating non-Markovian effects and for sampling. We conclude by identifying opportunities for further research, including connections to reinforcement learning and generative modeling.
Related
- The Thiolate Shell as a Dynamic Gold Reservoir: Size-Dependent Thermal Restructuring of Gold Nanoclusters from MACE-Based Machine-Learning Molecular Dynamics
- Machine-Learning Molecular Dynamics Reveals Grotthuss Transport in Concentrated Electrolytes
- Unveiling Phase Transformation Mechanism in BN from Large-Scale Machine Learning Molecular Dynamics
- Insights into Desalination Performance of Spacing-Tunable Laminated MXene Membranes via Molecular Dynamics and Machine Learning
- Universal Machine-learning Molecular Dynamics at the Speed of Empirical Potentials
BibTeX
@misc{weare2026machine,
title = {{Machine learning kinetics from molecular dynamics data}},
author = {Jonathan Weare and Aaron R. Dinner},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2609.17736},
eprint = {2609.17736},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.17736},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Machine learning kinetics from molecular dynamics data
AU - Jonathan Weare
AU - Aaron R. Dinner
PY - 2026
DA - 2026/10/25
PB - arXiv
DO - 10.48550/arxiv.2609.17736
UR - https://arxiv.org/abs/2609.17736
AB - Most molecular transitions occur on time scales far beyond direct molecular dynamics simulations. The committor, the probability that a configuration reaches a product state before a reactant state, is a central kinetic statistic, providing a mechanism-independent reaction coordinate and a foundation for transition path theory and the calculation of rates. This review surveys modern approaches for estimating the committor and related kinetic statistics from molecular simulations, with an emphasis on self-supervised methods that learn solutions of their defining dynamical equations rather than relying on labeled shooting data. We develop a common operator viewpoint connecting generator-based partial differential equations, variational principles, Markov state models, dynamical Galerkin approximation, and neural networks. Empirical and theoretical evidence points to the efficiency of these methods. We provide theoretical and practical guidance for realizing their full potential in applications, including strategies for treating non-Markovian effects and for sampling. We conclude by identifying opportunities for further research, including connections to reinforcement learning and generative modeling.
ER -